Congrats! Our GazeSAM work is accepted in NeurIPS 2023 Workshop Gaze Meets ML
The GazeSAM work introduces eye gaze as an interactive prompt for image segmentation and was accepted at the NeurIPS 2023 Gaze Meets ML workshop.
This study introduces eye gaze as a novel interactive prompt for image segmentation, different than previous model-based applications. Specifically, leveraging the real-time interactive prompting feature of the recently proposed Segment Anything Model (SAM), we present the GazeSAM system to enable users to collect target segmentation masks by simply looking at the region of interest.
Github Page: ukaukaaaa.github.io
paper link: openreview.net
